The Internet of Things (IoT) at a high level is relatively simple—you have sensors in IoT devices that take readings and communicate that data through the internet and into the cloud to another device or some type of analytics user interface. Essentially, data is collected and can then be interpreted into actionable insights.
Say you have a piece of equipment on the production line in a manufacturing plant with a sensor-based IoT device attached. This device is monitoring the health of that machine to make sure that the needs of the machine are met before damage or downtime occurs. Unplanned downtime is one of the biggest cost drivers in industrial manufacturing, posing a $50 billion annual threat to manufacturers.
These industrial IoT devices are a critical factor in a successful production run, so it stands to reason that manufacturers rely heavily on the ecosystem of these devices to work. It is increasingly likely that modern, cost-efficient IoT devices are replacing older technologies to measure all types of information within an industrial plant, such as temperature, to avoid overheating and damage to equipment and products, pressure in tanks that hold liquids and so forth.
This is what I like to call the visible part of the iceberg when it comes to IoT. The end user, which in this example is the manufacturer, is overseeing, managing and benefitting from their IoT-enabled solution represented by the part of the iceberg above the water line. But underneath the surface is so much more that supports what is seen above the surface. And this is where all the complexity comes in.
The Rest Of The Iceberg
What is under the surface—the other 80% to 90% of the iceberg—are all the intricate complexities of IoT that drive an IoT ecosystem’s operations and successes. It begins with the device itself. That device needs a SIM card—whether that is a traditional SIM or an eSIM—which needs to be activated on a carrier network, and the device needs to be configured to the network. This process allows devices to work out of the box when they arrive as part of a deployment. Then a gateway or router is required, which needs to be device and network compatible.
All hardware needs to be kitted and shipped in a manner that is aligned with the organization’s deployment. If devices or hardware are coming from multiple OEMs or wholesalers, then this can get relatively complicated rapidly. As the IoT infrastructure is built (devices, networks and applications), more and more touchpoints are added, and each requires a certain number of tasks and considerations so that everything properly works.
What I have just described is mostly concerned with the initial deployment of an IoT solution. The management of IoT can be just as complex. Imagine that you are an OEM, and you have 1,000 IoT devices on your manufacturing floor helping you manage operations for five different product lines or devices. Each line has 200 devices helping to monitor the health of the production line, and each group has its own lifecycle, which might be anywhere between two and 10 years.
What happens when one group of devices reaches the end of a lifecycle? Suddenly the carousel of device management and logistics never stops turning, and it is a constant effort. You must deploy new devices using the same process of ordering, activating, configuring, kitting and shipping as before. It starts to get complicated just running the logistics of keeping the right number of devices online and working.
We could further complicate this scenario by including any regulatory compliance to which this OEM must adhere when introducing devices into operations or data collection and storage. This is certainly true when you consider IoT in connected health deployments, which have regulatory compliance intricately woven into the many aspects of patient data collection and transmission.
Or if you want to deploy solutions globally, then devices, network communications, data storage and so on fall under different regulations and compliance as your IoT deployment of origin. And, of course, different connectivity protocols (or even if the deployment is entirely in LTE) mean different MNOs are required for resilient, high-quality local connectivity.
All of this is to say that IoT solutions can get very complicated, and that can hinder success. As previously stated, the end user is interested in the tip of the iceberg and might not always have the time, resources or finances to manage the under-the-surface details that support success.
Tackling Complexities Below The Surface
Beecham Research famously published a study a few years ago, “Why IoT Projects Fail,” where only 26% of those surveyed reported being successful with their IoT initiatives. That amounts to a pretty high failure rate. While there are many factors at play, including some objectivity on what might be considered a success, a lot of the struggles and issues listed in the study fall under the umbrella of IoT-managed services.
That being said, there are several approaches to mitigating these types of issues, whether it is creating a department or sub-department within the organization or its IT infrastructure, pursuing software or other technology applications, or looking outside of the organization to a third party or coined IoT professional service that focuses on managed services and acts as an extension of the organization.
IoT can provide incredible benefits, and I believe it is going to become an intrinsic part of business operations across all industries, but those below-the-surface details and complexities must be addressed and managed for success.
